CN102521052A - Resource allocation method of virtualized data center and virtual machine monitor - Google Patents
Resource allocation method of virtualized data center and virtual machine monitor Download PDFInfo
- Publication number
- CN102521052A CN102521052A CN2011104016202A CN201110401620A CN102521052A CN 102521052 A CN102521052 A CN 102521052A CN 2011104016202 A CN2011104016202 A CN 2011104016202A CN 201110401620 A CN201110401620 A CN 201110401620A CN 102521052 A CN102521052 A CN 102521052A
- Authority
- CN
- China
- Prior art keywords
- virtual machine
- resource
- server
- alternative
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 32
- 238000013468 resource allocation Methods 0.000 title claims abstract description 8
- 230000001186 cumulative effect Effects 0.000 claims abstract description 9
- 238000005315 distribution function Methods 0.000 claims abstract description 9
- 230000005012 migration Effects 0.000 claims description 14
- 238000013508 migration Methods 0.000 claims description 14
- 229920000642 polymer Polymers 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 2
- 239000010410 layer Substances 0.000 description 13
- 238000005516 engineering process Methods 0.000 description 11
- 238000004891 communication Methods 0.000 description 6
- 239000012792 core layer Substances 0.000 description 5
- 238000010586 diagram Methods 0.000 description 4
- 230000003993 interaction Effects 0.000 description 4
- 201000004569 Blindness Diseases 0.000 description 2
- 241001269238 Data Species 0.000 description 2
- 238000012797 qualification Methods 0.000 description 2
- 239000012141 concentrate Substances 0.000 description 1
- 238000007726 management method Methods 0.000 description 1
- 238000006116 polymerization reaction Methods 0.000 description 1
Images
Landscapes
- Computer And Data Communications (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
本发明提供一种虚拟化数据中心资源分配方法及虚拟机监控器，其中方法包括：虚拟机监控器根据接收到的用户应用请求中所请求构建的虚拟机，选择用于运行所述虚拟机的执行服务器；在所述执行服务器上为各个所述虚拟机分配能够满足该虚拟机的资源需求量的初始资源，并基于该初始资源构建所述虚拟机；在所述虚拟机的预设运行时段内统计该虚拟机的历史资源消耗量；根据所述历史资源消耗量及所述资源需求量确定超售比率O，其中，O∈[0，100]；在所述执行服务器上为各个所述虚拟机分配最终资源，该最终资源＝A(O)，并基于该最终资源重新构建所述虚拟机，其中，A为所述历史资源消耗量的累积分布函数。本发明减少了资源的过量分配，提高了资源利用率。
The present invention provides a method for allocating resources in a virtualized data center and a virtual machine monitor, wherein the method includes: the virtual machine monitor selects a virtual machine for running the virtual machine according to the virtual machine requested in the received user application request Executing the server; assigning initial resources capable of meeting the resource requirements of the virtual machine to each virtual machine on the executing server, and constructing the virtual machine based on the initial resources; during the preset running period of the virtual machine Internally count the historical resource consumption of the virtual machine; determine the overbooking ratio O according to the historical resource consumption and the resource demand, where O∈[0,100]; The virtual machine allocates a final resource, where the final resource=A(O), and rebuilds the virtual machine based on the final resource, where A is a cumulative distribution function of the historical resource consumption. The invention reduces excessive resource allocation and improves resource utilization.
Description
Technical field
The present invention relates to a kind of virtual data center resource allocation methods and monitor of virtual machine, belong to the resource allocation techniques field of virtual data center.
Background technology
Intel Virtualization Technology is a kind of concurrent and technology of independent operating on identical hardware resource of a plurality of operation system examples that allow.This technological core is to utilize monitor of virtual machine unified management and these virtual machines of scheduling on the hardware resource that respective service provides, to carry out.Utilize Intel Virtualization Technology, will use individual packages in virtual machine,, therefore allow the application of a plurality of isomeries to operate in same station server, thereby realized the polymerization of using owing to isolate each other between the virtual machine.
The existing data center that has used Intel Virtualization Technology concentrates in together physical resource and forms a shared virtual resource pond, and the user can use computational resource as required through the mode of leasing, and adopts the charge method of living within one's income.Yet; Existing Intel Virtualization Technology only carries out resources allocation according to user's request, and the user often has blindness when the application request resource, and they apply for resource at often maximum resource use amount; Final consumed resource when making the resources requirement of the virtual machine that the user asks to make up often be higher than the virtual machine operation; Thereby cause the excessive distribution of resource, make resource utilization low excessively, the more applications request can't be met.
Summary of the invention
The present invention provides a kind of virtual data center resource allocation methods and monitor of virtual machine, in order to reduce the excessive distribution of resource, improves resource utilization.
One aspect of the present invention provides a kind of virtual data center resource allocation methods, comprising:
Monitor of virtual machine selects to be used to move the execution server of said virtual machine according to the virtual machine of asking in the user's application request that receives to make up;
Be the initial resource that each said virtual machine distributes the resources requirement that can satisfy this virtual machine on said execution server, and make up said virtual machine based on this initial resource;
The historical consumed resource of this virtual machine of statistics in the preset operation period of said virtual machine;
Sell ratio O according to said historical consumed resource and definite surpassing of said resources requirement, wherein, O ∈ [0,100];
On said execution server, distribute final resource for each said virtual machine, this final resource=A (O), and rebuild said virtual machine based on this final resource, wherein, A is the cumulative distribution function of said historical consumed resource.
The present invention provides a kind of monitor of virtual machine on the other hand, comprising:
Server is selected module, is used for asking according to the user's application request that receives the virtual machine of structure, selects to be used to move the execution server of said virtual machine;
The original allocation module, being used on the said execution server that server selects module to select is the initial resource that each said virtual machine distributes the resources requirement that can satisfy this virtual machine, and makes up said virtual machine based on this initial resource;
The historical statistics module is used for the preset historical consumed resource that moves this virtual machine of statistics in the period at the said virtual machine of original allocation module construction;
Surpass and sell computing module, the said historical consumed resource and definite the surpassing of said resources requirement that are used for counting on according to the historical statistics module are sold ratio O, wherein, and O ∈ [0,100];
The final assignment module; Be used on the said execution server that server selects module to select, distributing final resource for each said virtual machine, this final resource=A (O), and rebuild said virtual machine based on this final resource; Wherein, A is the cumulative distribution function of said historical consumed resource.
The present invention is through reducing the resource distribute to virtual machine based on the ultra ratio O that sells that calculates, thereby under the prerequisite that guarantees the normal operation of virtual machine as far as possible, reduced the excessive distribution of resource, improved the utilization factor of resource.
Description of drawings
Fig. 1 is the process flow diagram of the virtual resource allocation methods embodiment of data center according to the invention;
Fig. 2 is the optional process flow diagram of step 100 shown in Figure 1;
Fig. 3 is the structural representation of monitor of virtual machine embodiment according to the invention;
Fig. 4 is the concrete structure synoptic diagram that server shown in Figure 3 is selected module 11.
Embodiment
Fig. 1 is the process flow diagram of the virtual resource allocation methods embodiment of data center according to the invention, and is as shown in the figure, and this method comprises the steps:
Step 100, monitor of virtual machine select to be used to move the execution server of said virtual machine according to the virtual machine of asking in the user's application request that receives to make up.
Wherein, said user's application request is to make up virtual machine so that realize concrete request of using by user's sent request, and virtual machine to be made up has one or more; Said execution server is used to that virtual machine to be made up provides resource so that the server of operation virtual machine.Concrete selection course can adopt greedy algorithm, will describe in subsequent content.
Step 200 is the initial resource that each said virtual machine distributes the resources requirement that can satisfy this virtual machine on said execution server, and makes up said virtual machine based on this initial resource.
Wherein, said initial resource is meant the resource of distributing according to user's demand, and the size of this initial resource can satisfy the resources requirement of each virtual machine.Of aforesaid background technology, user's request often has blindness, and therefore, this initial resource of being distributed often is greater than the actual consumed resource of virtual machine, and the part that wherein has more is called the ultra resource of selling in the present embodiment.
Wherein, be a period of time that virtual machine is moved based on initial resource in the said preset operation period, purpose is in order during this period of time to obtain the actual historical consumed resource that takies of virtual machine, so that the reference when selling resource as follow-up assessment is ultra.
Wherein, the said ultra ratio O that sells is meant the ultra resource request that resource will satisfy the user with the probability of (1-O) % of selling.Can be expressed as with mathematical expression: Pr (x＞A (O))＜=O%, wherein, A (O) expression is sold ratio O ultra resource of selling for this virtual machine distribution with ultra, and x representes the historical consumed resource of this virtual machine, and Pr representes probability.Ultra variation range of selling ratio O is between 0～100, and O=0 representes not ultra selling, and asks Resources allocation according to the user fully; During O=100, then Resources allocation is zero.
Said ultra when selling ratio O when confirming, can consider that actual gain finds the solution following equation:
E(R|O)＝Req×price-A(O)×price-(∑Req)×cost×Pr(x
＞(A(O)+A′))
Wherein, E (R|O) is illustrated in and surpasses the prospective earnings value that makes up a virtual machine under the condition of selling ratio O, and Req representes the resources requirement of this virtual machine; Price representes resource income unit price; Can be according to service-level agreement (Service Level Agreement, abbreviation: confirm that SLA) A (O) expression is sold ratio O ultra resource of selling for this virtual machine distribution with ultra; It satisfies cumulative distribution function (the Cumulative Distribution Function of resource consumption probability; Be called for short: CDF), ∑ Req representes that this virtual machine belongs to the summation of the resources requirement of carrying out the whole virtual machines that move on server, and cost representes conflict reparations unit price; Can confirm according to SLA; X representes the historical consumed resource of this virtual machine, and this virtual machine place of A ' expression is carried out and is the final resource except that this virtual assigned virtual machine on the server, and Pr representes probability.
The practical significance of following formula is: the resource that is a virtual machine distribution is many more; Income is many more, but ultra sell resource and in fact be not fully utilized, and makes up more multi-dummy machine thereby hindered; Therefore deduct the ultra income that resource obtains of selling, also deduct the damages of paying when the virtual machine conflict takes place in addition.Relevant virtual machine conflict will be introduced in subsequent content.Concrete equation solution process can adopt process of iteration to realize, below lifts an example explanation:
Wherein, Said final resource is meant from initial resource removes the ultra resource that final assignment is given virtual machine after the resource of selling; A is the cumulative distribution function of said historical consumed resource, owing to saidly ultra sell the possibility that resource has only O% and can be utilized, therefore; Do not sell resources allocation to said virtual machine even this part is not surpassed, can not make a big impact yet it.
The said method of present embodiment is through reducing the resource distribute to virtual machine based on the ultra ratio O that sells that calculates, thereby under the prerequisite that guarantees the normal operation of virtual machine as far as possible, reduced the excessive distribution of resource, improved the utilization factor of resource.
In addition; The said ultra possibility of selling resource O% still being arranged of present embodiment can be utilized; When virtual machine has utilized this part ultra when selling resource, a real resource consumption of carrying out the whole virtual machines that move on the server might surpass the available volume of resources that this executions server can provide, and just the virtual machine conflict can take place at this moment; That is to say that the virtual machine that is distributed has exceeded the ability of place execution server, can continue to carry out following step 600 this moment.
Step 600 when clashing between a plurality of virtual machines on the said execution server, is carried out the virtual machine (vm) migration that clashes another in same alternative group on server.
Particularly, virtual machine (vm) migration that can committed memory space in the virtual machine that clash is minimum is carried out on the server to said another, so that reduce the expense that migration is produced.
Following with reference to Fig. 2, specify and adopt greedy algorithm in the above-mentioned steps 100 carrying out the selection course of server, as shown in Figure 2, comprise the steps:
Existing Intel Virtualization Technology adopts three layers of tree topology structure usually: top layer is core layer (core level), is used to realize concrete application; The middle level is a polymer layer, includes many switches, is used to realize exchanges data; Bottom is an Access Layer, comprises multiple servers, is used to provide resource, like the computing power of CPU, memory headroom etc., to make up virtual machine.Owing to can move a plurality of virtual machines on the station server, so its available resources are dynamic changes, need carry out by the time collect.
Wherein, said alternative services device is meant to have the server that enough resources are provided for virtual machine to be made up, if enough resources can not be provided, then can not be chosen as the alternative services device.
In the prior art; Comprise a plurality of switches in the polymer layer; Each switch directly links to each other with a plurality of servers; In this step, will be divided into one alternative group with the alternative services device that same switch in the polymer layer directly links to each other, will be divided into the alternative services device that another switch directly links to each other another alternative group (pod), the alternative services device in different alternative group will connect via the switch indirect communication.
Suppose that each virtual machine all needs the memory headroom of 2M; Alternative services device in first alternative group can provide the available memory space of 10M altogether; Alternative services device in second alternative group can provide the available memory space of 4M altogether, and the alternative services device in the 3rd alternative group can provide the available memory space of 6M altogether, and the alternative services device in the 4th alternative group can provide the available memory space of 2M altogether; Then through the statistics that obtains after this step be: first alternative group can make up 5 virtual machines at most; Second alternative group can make up 2 virtual machines at most, and the 3rd alternative group can make up 3 virtual machines at most, and the 4th alternative group can make up 1 virtual machine at most.
Need to prove that here in fact, the resources requirement of different virtual machine possibly have nothing in common with each other, and only illustrates, and does not do qualification here.
For example; In last example; Suppose that the virtual machine that the user asks to make up adds up to 10; Then the selection course to alternative group is in this step: select first alternative group earlier so that make up 5 virtual machines, select the 3rd alternative group then so that make up 3 virtual machines, select second alternative group at last then so that make up 2 virtual machines.
Through method shown in Figure 2; Three layers of tree topology structure of existing Intel Virtualization Technology when selecting to carry out server, have been considered for virtual machine; Virtual machine is placed on as far as possible in alternative group of the execution server that comprises maximum quantity; Because the execution server in alternative group all directly links to each other with same switch; Therefore operate in these communication datas of carrying out between the virtual machine on the server and need not forwarding, and reduced the communication interaction between the virtual machine in different alternative group as far as possible, thereby help reducing the burden of core layer via core layer.This method also can be called greedy algorithm.
Need to prove here; Because the execution server in selected each alternative group all directly links to each other with an exchange; Therefore concrete which virtual machine of operation is not having difference in essence aspect the communication data forwarding amount on concrete which execution server, does not do qualification here.
Fig. 3 is the structural representation of monitor of virtual machine embodiment according to the invention; In order to realize said method; As shown in the figure, this monitor of virtual machine 10 comprises at least: server is selected module 11, original allocation module 12, historical statistics module 13, ultra computing module 14 and the final assignment module 15 of selling.Need to prove that here the thick vertical line among the figure is represented bus, each template all is connected on the bus, but in the present embodiment, as long as can make above-mentioned each module realize necessary information interaction, adopts other connected modes also to be fine.Below introduce the principle of work of monitor of virtual machine 10 in detail:
When monitor of virtual machine 10 receives user's application request, select module 11 according to the virtual machine of asking in the user's application request that receives to make up by server, select to be used to move the execution server of said virtual machine; Be the initial resource that each said virtual machine distribution can be satisfied the resources requirement of this virtual machine on the said execution server that server selects module 11 to select by original allocation module 12 then, and make up said virtual machine based on this initial resource.
After this, in the preset operation period of the said virtual machine that original allocation module 12 makes up, add up the historical consumed resource of this virtual machine by historical statistics module 13; Wherein, be a period of time that virtual machine is moved based on initial resource in the said preset operation period, purpose is in order during this period of time to obtain the actual historical consumed resource that takies of virtual machine, so that the reference when selling resource as follow-up assessment is ultra.
After this, sell ratio O according to said historical consumed resource and definite surpassing of said resources requirement that historical statistics module 13 counts on by the ultra computing module 14 of selling, wherein, O ∈ [0,100].Said ultra when selling ratio O when confirming, can consider that actual gain finds the solution following equation:
E(R|O)＝Req×price-A(O)×price-(∑Req)×cost×Pr(x
＞(A(O)+A′))
Wherein relevant ultra detailed description of selling each several part in ratio O and the above-mentioned equation can be repeated no more referring to said method embodiment here.
At last; On the said execution server that server selects module 11 to select, distribute final resource by final assignment module 15 for each said virtual machine; This final resource=A (O); And rebuild said virtual machine based on this final resource, wherein, A is the cumulative distribution function of said historical consumed resource.
The said method of present embodiment is through reducing the resource distribute to virtual machine based on the ultra ratio O that sells that calculates, thereby under the prerequisite that guarantees the normal operation of virtual machine as far as possible, reduced the excessive distribution of resource, improved the utilization factor of resource.
In addition; As shown in Figure 3; When server is selected to clash between a plurality of virtual machines on the said execution server that module 11 selects, virtual machine (vm) migration module 16 can further be set, the virtual machine (vm) migration that clashes another in same alternative group carried out on server.Particularly, virtual machine (vm) migration that can committed memory space in the virtual machine that clash is minimum is carried out on the server to said another, so that reduce the expense that migration is produced.
Following with reference to Fig. 4, specify the inner structure that said server is selected module 11, as shown in the figure, this server selects module 11 to comprise at least: collector unit 1101, alternative unit 1102, division unit 1103, statistic unit 1104 and selected cell 1105.Need to prove that here the thick vertical line among the figure is represented bus, each template all is connected on the bus, but in the present embodiment, as long as can make above-mentioned each module realize necessary information interaction, adopts other connected modes also to be fine.Below the explanation server is selected the principle of work of module 11:
The available resource information of collecting the whole servers that are positioned at Access Layer by collector unit 1101 earlier; The available resource information of being collected according to collector unit 1101 by alternative unit 1102 then can make up the server of said virtual machine and confirm as the alternative services device; Wherein, said alternative services device is meant to have the server that enough resources are provided for virtual machine to be made up, if enough resources can not be provided, then can not be chosen as the alternative services device.
After this, to the said alternative services device of confirming by alternative unit 1102, will be divided into one alternative group with the alternative services device that same switch in the polymer layer directly links to each other by division unit 1103.The detailed description of three layers of tree topology structure of relevant Intel Virtualization Technology can be repeated no more referring to said method embodiment here.
After this, add up the quantity of each said alternative group of virtual machine that can make up at most of dividing by division unit 1103 according to the resource requirement information of each said virtual machine by statistic unit 1104; And select the respective alternative group successively according to the said quantity order from big to small that statistic unit 1104 counts on by selected cell 1105; Equal the sum of virtual machine to be made up up to the summation of selected alternative group of corresponding said quantity, and the alternative services device in selected said alternative group is chosen as said execution server.
Through structure shown in Figure 4; Three layers of tree topology structure of existing Intel Virtualization Technology when selecting to carry out server, have been considered for virtual machine; Virtual machine is placed on as far as possible in alternative group of the execution server that comprises maximum quantity; Because the execution server in alternative group all directly links to each other with same switch; Therefore operate in these communication datas of carrying out between the virtual machine on the server and need not forwarding, and reduced the communication interaction between the virtual machine in different alternative group as far as possible, thereby help reducing the burden of core layer via core layer.This method also can be called greedy algorithm.
One of ordinary skill in the art will appreciate that: all or part of step that realizes said method embodiment can be accomplished through the relevant hardware of programmed instruction; Aforesaid program can be stored in the computer read/write memory medium; This program the step that comprises said method embodiment when carrying out; And aforesaid storage medium comprises: various media that can be program code stored such as ROM, RAM, magnetic disc or CD.
What should explain at last is: above embodiment is only in order to explaining technical scheme of the present invention, but not to its restriction; Although with reference to previous embodiment the present invention has been carried out detailed explanation, those of ordinary skill in the art is to be understood that: it still can be made amendment to the technical scheme that aforementioned each embodiment put down in writing, and perhaps part technical characterictic wherein is equal to replacement; And these are revised or replacement, do not make the spirit and the scope of the essence disengaging various embodiments of the present invention technical scheme of relevant art scheme.
Claims (10)
1. a virtual data center resource allocation methods is characterized in that, comprising:
Monitor of virtual machine selects to be used to move the execution server of said virtual machine according to the virtual machine of asking in the user's application request that receives to make up;
Be the initial resource that each said virtual machine distributes the resources requirement that can satisfy this virtual machine on said execution server, and make up said virtual machine based on this initial resource;
The historical consumed resource of this virtual machine of statistics in the preset operation period of said virtual machine;
Sell ratio O according to said historical consumed resource and definite surpassing of said resources requirement, wherein, O ∈ [0,100];
On said execution server, distribute final resource for each said virtual machine, this final resource=A (O), and rebuild said virtual machine based on this final resource, wherein, A is the cumulative distribution function of said historical consumed resource.
2. method according to claim 1 is characterized in that, said monitor of virtual machine selects said server to comprise:
Collection is positioned at the available resource information of whole servers of Access Layer;
The server that can make up said virtual machine according to the available resource information of collecting is confirmed as the alternative services device;
To be divided into one alternative group with the alternative services device that same switch in the polymer layer directly links to each other;
Add up the quantity of each said alternative group of virtual machine that can make up at most according to the resource requirement information of each said virtual machine;
Said quantity order from big to small according to counting on is selected the respective alternative group successively, equals the sum of virtual machine to be made up up to the summation of selected alternative group of corresponding said quantity;
Alternative services device in selected said alternative group is chosen as said execution server.
3. method according to claim 1 and 2 is characterized in that, saidly confirms ultraly to sell ratio O and comprise according to said historical consumed resource: according to the value of the said O of following Equation for Calculating,
E(R|O)＝Req×price-A(O)×price-(∑Req)×cost×Pr(x
＞(A(O)+A′))
Wherein, E (R|O) is illustrated in and surpasses the prospective earnings value that makes up a virtual machine under the condition of selling ratio O, and Req representes the resources requirement of this virtual machine, and price representes resource income unit price; A (O) expression is sold ratio O ultra resource of selling for this virtual machine distribution with ultra; ∑ Req representes that this virtual machine belongs to the summation of the resources requirement of carrying out the whole virtual machines that move on server, and cost representes conflict reparations unit price, and x representes the historical consumed resource of this virtual machine; This virtual machine place of A ' expression is carried out and is the final resource except that this virtual assigned virtual machine on the server, and Pr representes probability.
4. method according to claim 3 is characterized in that, also comprises: when clashing between a plurality of virtual machines on the said execution server, the virtual machine (vm) migration that clashes another in same alternative group carried out on server.
5. method according to claim 4; It is characterized in that the said virtual machine (vm) migration that clashes is carried out on the server to said another comprises: the virtual machine (vm) migration that committed memory space in the virtual machine that clashes is minimum is carried out on the server to said another.
6. a monitor of virtual machine is characterized in that, comprising:
Server is selected module, is used for asking according to the user's application request that receives the virtual machine of structure, selects to be used to move the execution server of said virtual machine;
The original allocation module, being used on the said execution server that server selects module to select is the initial resource that each said virtual machine distributes the resources requirement that can satisfy this virtual machine, and makes up said virtual machine based on this initial resource;
The historical statistics module is used for the preset historical consumed resource that moves this virtual machine of statistics in the period at the said virtual machine of original allocation module construction;
Surpass and sell computing module, the said historical consumed resource and definite the surpassing of said resources requirement that are used for counting on according to the historical statistics module are sold ratio O, wherein, and O ∈ [0,100];
The final assignment module; Be used on the said execution server that server selects module to select, distributing final resource for each said virtual machine, this final resource=A (O), and rebuild said virtual machine based on this final resource; Wherein, A is the cumulative distribution function of said historical consumed resource.
7. monitor of virtual machine according to claim 6 is characterized in that, said server selects module to comprise:
Collector unit, the available resource information that is used to collect the whole servers that are positioned at Access Layer;
Alternative unit, the available resource information that is used for collecting according to collector unit can make up the server of said virtual machine and confirm as the alternative services device;
Division unit is used for will being divided into one alternative group with the alternative services device that same switch in the polymer layer directly links to each other to by the definite said alternative services device in alternative unit;
Statistic unit is used for adding up the quantity by each said alternative group of virtual machine that can make up at most of division unit division according to the resource requirement information of each said virtual machine;
Selected cell; The said quantity order from big to small that is used for counting on according to statistic unit is selected the respective alternative group successively; Equal the sum of virtual machine to be made up up to the summation of selected alternative group of corresponding said quantity, and the alternative services device in selected said alternative group is chosen as said execution server.
8. according to claim 6 or 7 described monitor of virtual machine, it is characterized in that the said ultra computing module of selling is used for the value according to the following said O of Equation for Calculating,
E(R|O)＝Req×price-A(O)×price-(∑Req)×cost×Pr(x
＞(A(O)+A′))
Wherein, E (R|O) is illustrated in and surpasses the prospective earnings value that makes up a virtual machine under the condition of selling ratio O, and Req representes the resources requirement of this virtual machine, and price representes resource income unit price; A (O) expression is sold ratio O ultra resource of selling for this virtual machine distribution with ultra; ∑ Req representes that this virtual machine belongs to the summation of the resources requirement of carrying out the whole virtual machines that move on server, and cost representes conflict reparations unit price, and x representes the historical consumed resource of this virtual machine; This virtual machine place of A ' expression is carried out and is the final resource except that this virtual assigned virtual machine on the server, and Pr representes probability.
9. monitor of virtual machine according to claim 8; It is characterized in that; Also comprise: the virtual machine (vm) migration module; Be used for when server is selected to clash between a plurality of virtual machines on the said execution server that module selects, the virtual machine (vm) migration that clashes another in same alternative group carried out on server.
10. monitor of virtual machine according to claim 9; It is characterized in that: said virtual machine (vm) migration module is used for when clashing between a plurality of virtual machines on the said execution server that server selection module is selected, and the virtual machine (vm) migration that committed memory space in the virtual machine that clashes is minimum is carried out on server to said another.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110401620 CN102521052B (en) | 2011-12-06 | 2011-12-06 | Resource allocation method of virtualized data center and virtual machine monitor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN 201110401620 CN102521052B (en) | 2011-12-06 | 2011-12-06 | Resource allocation method of virtualized data center and virtual machine monitor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN102521052A true CN102521052A (en) | 2012-06-27 |
| CN102521052B CN102521052B (en) | 2013-12-25 |
Family
ID=46291987
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN 201110401620 Expired - Fee Related CN102521052B (en) | 2011-12-06 | 2011-12-06 | Resource allocation method of virtualized data center and virtual machine monitor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN102521052B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103336722A (en) * | 2013-07-16 | 2013-10-02 | 上海大学 | Virtual machine CPU source monitoring and dynamic distributing method |
| CN103634330A (en) * | 2012-08-20 | 2014-03-12 | 曙光信息产业(北京)有限公司 | Automatic resource distribution method in cloud calculation environment |
| GB2508161A (en) * | 2012-11-21 | 2014-05-28 | Ibm | Monitoring applications executing on a virtual machine and allocating the required resources to the virtual machine. |
| WO2015085475A1 (en) * | 2013-12-10 | 2015-06-18 | 华为技术有限公司 | Method and apparatus for scheduling virtual machine resources |
| CN105320559A (en) * | 2014-07-30 | 2016-02-10 | 中国移动通信集团广东有限公司 | Scheduling method and device of cloud computing system |
| CN105843678A (en) * | 2016-03-17 | 2016-08-10 | 国网山东省电力公司信息通信公司 | Optimization theory-based theory and system for scheduling and migrating resources of virtual machine |
| CN107025137A (en) * | 2016-11-24 | 2017-08-08 | 阿里巴巴集团控股有限公司 | A kind of resource query method and device |
| CN108196958A (en) * | 2017-12-29 | 2018-06-22 | 北京泽塔云科技股份有限公司 | Scheduling of resource distribution method, computer system and super fusion architecture system |
| CN105335208B (en) * | 2014-06-18 | 2019-01-25 | 上海诺基亚贝尔股份有限公司 | Determine the method and apparatus of physical computing resources collision probability in virtual environment |
| CN111124657A (en) * | 2018-10-31 | 2020-05-08 | 北京金山云网络技术有限公司 | Resource management method, device, electronic equipment and storage medium |
| CN113672392A (en) * | 2021-08-25 | 2021-11-19 | 国家电网有限公司信息通信分公司 | Method and apparatus for energy consumption optimization, electronic device and computer-readable storage medium |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050262504A1 (en) * | 2004-05-21 | 2005-11-24 | Esfahany Kouros H | Method and apparatus for dynamic CPU resource management |
| CN101488098A (en) * | 2009-02-13 | 2009-07-22 | 华中科技大学 | Multi-core computing resource management system based on virtual computing technology |
| CN101593134A (en) * | 2009-06-29 | 2009-12-02 | 北京航空航天大学 | Virtual machine CPU resource allocation method and device |
| US20100199285A1 (en) * | 2009-02-05 | 2010-08-05 | Vmware, Inc. | Virtual machine utility computing method and system |
-
2011
- 2011-12-06 CN CN 201110401620 patent/CN102521052B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050262504A1 (en) * | 2004-05-21 | 2005-11-24 | Esfahany Kouros H | Method and apparatus for dynamic CPU resource management |
| US20100199285A1 (en) * | 2009-02-05 | 2010-08-05 | Vmware, Inc. | Virtual machine utility computing method and system |
| CN101488098A (en) * | 2009-02-13 | 2009-07-22 | 华中科技大学 | Multi-core computing resource management system based on virtual computing technology |
| CN101593134A (en) * | 2009-06-29 | 2009-12-02 | 北京航空航天大学 | Virtual machine CPU resource allocation method and device |
Cited By (17)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103634330A (en) * | 2012-08-20 | 2014-03-12 | 曙光信息产业(北京)有限公司 | Automatic resource distribution method in cloud calculation environment |
| GB2508161A (en) * | 2012-11-21 | 2014-05-28 | Ibm | Monitoring applications executing on a virtual machine and allocating the required resources to the virtual machine. |
| US9483288B2 (en) | 2012-11-21 | 2016-11-01 | International Business Machines Corporation | Method and system for running a virtual appliance |
| CN103336722B (en) * | 2013-07-16 | 2017-09-26 | 上海大学 | A kind of CPU resources of virtual machine monitoring and dynamic allocation method |
| CN103336722A (en) * | 2013-07-16 | 2013-10-02 | 上海大学 | Virtual machine CPU source monitoring and dynamic distributing method |
| WO2015085475A1 (en) * | 2013-12-10 | 2015-06-18 | 华为技术有限公司 | Method and apparatus for scheduling virtual machine resources |
| CN105378667A (en) * | 2013-12-10 | 2016-03-02 | 华为技术有限公司 | Method and apparatus for scheduling virtual machine resources |
| CN105378667B (en) * | 2013-12-10 | 2018-09-28 | 华为技术有限公司 | A kind of dispatching method and device of resources of virtual machine |
| CN105335208B (en) * | 2014-06-18 | 2019-01-25 | 上海诺基亚贝尔股份有限公司 | Determine the method and apparatus of physical computing resources collision probability in virtual environment |
| CN105320559A (en) * | 2014-07-30 | 2016-02-10 | 中国移动通信集团广东有限公司 | Scheduling method and device of cloud computing system |
| CN105320559B (en) * | 2014-07-30 | 2019-02-19 | 中国移动通信集团广东有限公司 | A scheduling method and device for a cloud computing system |
| CN105843678A (en) * | 2016-03-17 | 2016-08-10 | 国网山东省电力公司信息通信公司 | Optimization theory-based theory and system for scheduling and migrating resources of virtual machine |
| CN107025137A (en) * | 2016-11-24 | 2017-08-08 | 阿里巴巴集团控股有限公司 | A kind of resource query method and device |
| CN108196958A (en) * | 2017-12-29 | 2018-06-22 | 北京泽塔云科技股份有限公司 | Scheduling of resource distribution method, computer system and super fusion architecture system |
| CN111124657A (en) * | 2018-10-31 | 2020-05-08 | 北京金山云网络技术有限公司 | Resource management method, device, electronic equipment and storage medium |
| CN111124657B (en) * | 2018-10-31 | 2024-03-08 | 北京金山云网络技术有限公司 | Resource management method, device, electronic equipment and storage medium |
| CN113672392A (en) * | 2021-08-25 | 2021-11-19 | 国家电网有限公司信息通信分公司 | Method and apparatus for energy consumption optimization, electronic device and computer-readable storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN102521052B (en) | 2013-12-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN102521052A (en) | Resource allocation method of virtualized data center and virtual machine monitor | |
| US10089147B2 (en) | High performance computing as a service | |
| Liu et al. | Job scheduling model for cloud computing based on multi-objective genetic algorithm | |
| US9542223B2 (en) | Scheduling jobs in a cluster by constructing multiple subclusters based on entry and exit rules | |
| CN102780759A (en) | Cloud computing resource scheduling method based on scheduling object space | |
| CN110597639B (en) | CPU distribution control method, device, server and storage medium | |
| CN110413201A (en) | For managing the method, equipment and computer program product of storage system | |
| CN103095788A (en) | Cloud resource scheduling policy based on network topology | |
| CN103180830A (en) | Cpu resource distribution method and calculating node | |
| CN106951330A (en) | A kind of maximized virtual machine distribution method of cloud service center service utility | |
| CN111538586A (en) | Cluster GPU resource management scheduling system, method and computer readable storage medium | |
| Chen et al. | Improving resource utilization via virtual machine placement in data center networks | |
| CN106412124B (en) | A kind of and sequence cloud service platform task distribution system and method for allocating tasks | |
| CN105373432A (en) | Cloud computing resource scheduling method based on virtual resource state prediction | |
| CN109976907A (en) | Method for allocating tasks and system, electronic equipment, computer-readable medium | |
| Ahmad et al. | An energy-efficient big data workflow scheduling algorithm under budget constraints for heterogeneous cloud environment | |
| US8060610B1 (en) | Multiple server workload management using instant capacity processors | |
| CN108073457A (en) | A kind of hierarchical resource management method of super fusion architecture, apparatus and system | |
| Taghinezhad-Niar et al. | Workflow scheduling of scientific workflows under simultaneous deadline and budget constraints | |
| CN118656177B (en) | A container scheduling method, device, medium and equipment based on GPU topology structure | |
| Mylavarapu et al. | An optimized capacity planning approach for virtual infrastructure exhibiting stochastic workload | |
| CN111796932A (en) | GPU resource scheduling method | |
| CN111626528A (en) | Bezier curve-based resource quota determining method and device and electronic equipment | |
| Xu et al. | Availability-constrained application deployment in hybrid cloud-edge collaborative environment | |
| Kiessler et al. | Optimization heuristics for cost-efficient long-term cloud portfolio allocations |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| C14 | Grant of patent or utility model | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20131225 Termination date: 20171206 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |